Brief Report: Estimating Differences and Ratios in Median Times to Event.
Ontology highlight
ABSTRACT:
SUBMITTER: Rogawski ET
PROVIDER: S-EPMC5039102 | biostudies-literature | 2016 Nov
REPOSITORIES: biostudies-literature
Publications
Brief Report: Estimating Differences and Ratios in Median Times to Event.
Rogawski Elizabeth T ET Westreich Daniel J DJ Kang Gagandeep G Ward Honorine D HD Cole Stephen R SR
Epidemiology (Cambridge, Mass.) 20161101 6
Time differences and time ratios are often more interpretable estimates of effect than hazard ratios for time-to-event data, especially for common outcomes. We developed a SAS macro for estimating time differences and time ratios between baseline-fixed binary exposure groups based on inverse probability-weighted Kaplan-Meier curves. The macro uses pooled logistic regression to calculate inverse probability of censoring and exposure weights, draws Kaplan-Meier curves based on the weighted data, a ...[more]
Similar Datasets
Project description:Adult sex ratio (ASR, the proportion of males in the adult population) is a central concept in population and evolutionary biology, and is also emerging as a major factor influencing mate choice, pair bonding and parental cooperation in both human and non-human societies. However, estimating ASR is fraught with difficulties stemming from the effects of spatial and temporal variation in the numbers of males and females, and detection/capture probabilities that differ between the sexes. Here, we critically evaluate methods for estimating ASR in wild animal populations, reviewing how recent statistical advances can be applied to handle some of these challenges. We review methods that directly account for detection differences between the sexes using counts of unmarked individuals (observed, trapped or killed) and counts of marked individuals using mark-recapture models. We review a third class of methods that do not directly sample the number of males and females, but instead estimate the sex ratio indirectly using relationships that emerge from demographic measures, such as survival, age structure, reproduction and assumed dynamics. We recommend that detection-based methods be used for estimating ASR in most situations, and point out that studies are needed that compare different ASR estimation methods and control for sex differences in dispersal.This article is part of the themed issue 'Adult sex ratios and reproductive decisions: a critical re-examination of sex differences in human and animal societies'.
| S-EPMC5540855 | biostudies-literature
Project description:ImportanceMedian organ waiting times published by transplant organizations may be biased when not appropriately accounting for censoring, death, and competing events. This can lead to overly optimistic waiting times for all transplant programs and, consequently, may deceive patients on the waiting list, transplant physicians, and health care policymakers.ObjectiveTo apply competing-risk multistate models to calculate probabilities for transplantation and adverse outcomes on the Swiss national transplant waiting list.Design, setting, and participantsThe WAIT (Waitlist Analysis in Transplantation) study was a retrospective cohort study of all transplant candidates in Switzerland listed from January 1, 2018, or later and observed until December 31, 2023. Transplant candidates were listed in 1 of the 6 transplant centers (Basel, Bern, Geneva, Lausanne, St Gallen, and Zurich) for heart, liver, lungs, kidney, or pancreas and/or islet transplant. A total of 4352 candidates were listed during the study period, of whom 709 (16.3%) were excluded due to living-donor transplant (691 in the kidney program and 18 in the liver program).ExposureWaiting for organ transplant.Main outcomes and measuresTime to transplantation, death, or delisting. Competing-risk multistate models were used to analyze time-to-event data from the national organ waiting list with the Aalen-Johansen estimator to compute probabilities for both transplant and adverse outcomes. Results were compared with the sample median among only those undergoing transplant and the Kaplan-Meier method with censoring of competing events.ResultsData from 3643 transplant candidates (2428 [66.6%] male; median age, 56 [range, 0-79] years) were included in the analysis. The median time to transplantation (MTT) was 0.91 (95% CI, 0.83-1.07) years for heart, 3.10 (95% CI, 2.57-3.77) years for kidney, 1.32 (95% CI, 0.76-1.55) years for liver, 0.80 (95% CI, 0.37-1.12) years for lung, and 1.62 (95% CI, 0.91-2.17) years for pancreas and/or islet programs. Alternative estimation methods introduced bias to varying degrees: the sample median among only persons undergoing transplantation underestimated the waiting time by 38% to 61% and the Kaplan-Meier method by 2% to 12% compared with the MTT.Conclusions and relevanceIn this cohort study of transplant candidates in Switzerland, the MTT, the duration at which the transplant probability is 0.50, was used as a measure of average waiting time. Suboptimal methods led to biased and overly optimistic waiting time estimations; thus, applying appropriate competing-risk methods to address censoring and competing events is crucial.
| S-EPMC11388028 | biostudies-literature
Project description:This paper deals with the analysis of recurrent event data subject to censored observation. Using a suitable adaptation of generalized estimating equations for longitudinal data, we propose a straightforward methodology for estimating the parameters indexing the conditional means and variances of the process interevent (i.e. gap) times. The proposed methodology permits the use of both time-fixed and time-varying covariates, as well as transformations of the gap times, creating a flexible and useful class of methods for analyzing gap-time data. Censoring is dealt with by imposing a parametric assumption on the censored gap times, and extensive simulation results demonstrate the relative robustness of parameter estimates even when this parametric assumption is incorrect. A suitable large-sample theory is developed. Finally, we use our methods to analyze data from a randomized trial of asthma prevention in young children.
| S-EPMC2697342 | biostudies-literature
Project description:BackgroundType 1 myocardial infarctions (T1MIs) result from atherosclerotic plaque instability, rupture, and/or erosion. Type 2 MIs (T2MIs) are secondary to causes such as sepsis and cocaine-induced vasospasm resulting in an oxygen demand-supply mismatch and are associated with higher mortality than T1MIs. T2MIs account for a higher proportion of MIs among people living with HIV (PLWH) compared with the general population. We compared MI rates by type among aging PLWH. We hypothesized that increases in MI rates with older age would differ by MI types, and T2MIs would be more common than T1MIs in younger individuals.MethodsPotential MIs from 6 sites were centrally adjudicated using physician notes, electrocardiograms, procedure results, and laboratory results. Reviewers categorized MIs by type and identified causes of T2MIs. We calculated T1MI and T2MI incidence rates. Incidence rate ratios were calculated for T2MI vs. T1MI rates per decade of age.ResultsWe included 462 T1MIs (52%) and 413 T2MIs (48%). T1MI rates increased with older age, although T1MIs occurred in all age decades including young adults. T2MI rates were significantly higher than T1MI rates for PLWH younger than 40 years. T1MI rates were similar or higher than T2MI rates among those older than 40 years (significantly higher for those aged 50-59 and 60-69 years).ConclusionsRates of T2MIs were higher than T1MIs until age 40 years among PLWH, differing from the general population, but rates of both were high among older PLWH. Given prognostic differences between MI types, these results highlight the importance of differentiating MI types among PLWH.
| S-EPMC8900222 | biostudies-literature
Project description:Molecular dating of species divergences has become an important means to add a temporal dimension to the Tree of Life. Increasingly larger datasets encompassing greater taxonomic diversity are becoming available to generate molecular timetrees by using sophisticated methods that model rate variation among lineages. However, the practical application of these methods is challenging because of the exorbitant calculation times required by current methods for contemporary data sizes, the difficulty in correctly modeling the rate heterogeneity in highly diverse taxonomic groups, and the lack of reliable clock calibrations and their uncertainty distributions for most groups of species. Here, we present a method that estimates relative times of divergences for all branching points (nodes) in very large phylogenetic trees without assuming a specific model for lineage rate variation or specifying any clock calibrations. The method (RelTime) performed better than existing methods when applied to very large computer simulated datasets where evolutionary rates were varied extensively among lineages by following autocorrelated and uncorrelated models. On average, RelTime completed calculations 1,000 times faster than the fastest Bayesian method, with even greater speed difference for larger number of sequences. This speed and accuracy will enable molecular dating analysis of very large datasets. Relative time estimates will be useful for determining the relative ordering and spacing of speciation events, identifying lineages with significantly slower or faster evolutionary rates, diagnosing the effect of selected calibrations on absolute divergence times, and estimating absolute times of divergence when highly reliable calibration points are available.
| S-EPMC3511068 | biostudies-literature
Project description:BackgroundSex differences in studies of antiretroviral (ART) drug exposure and treatment outcomes support the hypothesis that some ART combinations may not be well tolerated in women. We evaluated disparities in outcomes between men and women participating in ACTG A5288, an interventional strategy trial for individuals failing a protease inhibitor-based second-line ART regimen in low- and middle-income countries.MethodsParticipants were assigned to one of 4 cohorts (A-D) based on resistance profiles and ART history. Cohort A had no lopinavir/ritonavir (LPV/r) resistance and stayed on their second-line regimen, and cohorts B, C, and D had increasing resistance and accessed novel ART regimens. In this secondary analysis, we evaluated sex differences in the primary endpoint, HIV-1 RNA ≤200 copies/mL at week 48; confirmed virologic failure ≥1000 copies/mL (VF); and clinical outcomes and adverse events (intent-to-treat).ResultsWomen made up 258/545 (47%) of the study population. More women than men were assigned to cohort A. Median follow-up was 72 weeks. Fewer women than men had HIV-1 RNA ≤200 copies/mL at week 48: 39% vs. 49% in cohort A and 83% vs. 89% in cohorts B, C, and D combined. More women experienced VF, grade ≥3 signs and symptoms, but similar grade ≥3 diagnoses or laboratory abnormalities.ConclusionsMore women than men entered the study with a resistance profile suggesting that their second-line regimen could have been effective in maintaining virologic suppression. The more frequent occurrence of grade ≥3 signs and symptoms in women suggests that tolerability issues were under recognized in women on protease inhibitor-based therapy.
| S-EPMC7228852 | biostudies-literature
Project description:Memorizing time of an event may employ two processes (1) encoding of the absolute time of events within an episode, (2) encoding of its relative order. Here we study interaction between these two processes. We performed experiments in which one or several items were presented, after which participants were asked to report the time of occurrence of items. When a single item was presented, the distribution of reported times was quite wide. When two or three items were presented, the relative order among them strongly affected the reported time of each of them. Bayesian theory that takes into account the memory for the events order is compatible with the experimental data, in particular in terms of the effect of order on absolute time reports. Our results suggest that people do not deduce order from memorized time, instead people's memory for absolute time of events relies critically on memorized order of the events.
| S-EPMC8408261 | biostudies-literature
Project description:Accurately dating when the first bilaterally symmetrical animals arose is crucial to our understanding of early animal evolution. The earliest unequivocally bilaterian fossils are approximately 555 million years old. In contrast, molecular-clock analyses calibrated by using the fossil record of vertebrates estimate that vertebrates split from dipterans (Drosophila) approximately 900 million years ago (Ma). Nonetheless, comparative genomic analyses suggest that a significant rate difference exists between vertebrates and dipterans, because the percentage difference between the genomes of mosquito and fly is greater than between fish and mouse, even though the vertebrate divergence is almost twice that of the dipteran. Here we show that the dipteran rate of molecular evolution is similar to other invertebrate taxa (echinoderms and bivalve molluscs) but not to vertebrates, which significantly decreased their rate of molecular evolution with respect to invertebrates. Using a data set consisting of the concatenation of seven different amino acid sequences from 23 ingroup taxa (giving a total of 11 different invertebrate calibration points scattered throughout the bilaterian tree and across the Phanerozoic), we estimate that the last common ancestor of bilaterians arose somewhere between 573 and 656 Ma, depending on the value assigned to the parameter scaling molecular substitution rate heterogeneity. These results are in accord with the known fossil record and support the view that the Cambrian explosion reflects, in part, the diversification of bilaterian phyla.
| S-EPMC404080 | biostudies-literature
Project description:Accurate estimation of divergence times of soil bacteria that form nitrogen-fixing associations with most leguminous plants is challenging because of a limited fossil record and complexities associated with molecular clocks and phylogenetic diversity of root nodule bacteria, collectively called rhizobia. To overcome the lack of fossil record in bacteria, divergence times of host legumes were used to calibrate molecular clocks and perform phylogenetic analyses in rhizobia. The 16S rRNA gene and intergenic spacer region remain among the favored molecular markers to reconstruct the timescale of rhizobia. We evaluate the performance of the random local clock model and the classical uncorrelated lognormal relaxed clock model, in combination with four tree models (coalescent constant size, birth-death, birth-death incomplete sampling, and Yule processes) on rhizobial divergence time estimates. Bayes factor tests based on the marginal likelihoods estimated from the stepping-stone sampling analyses strongly favored the random local clock model in combination with Yule process. Our results on the divergence time estimation from 16S rRNA gene and intergenic spacer region sequences are compatible with age estimates based on the conserved core genes but significantly older than those obtained from symbiotic genes, such as nodIJ genes. This difference may be due to the accelerated evolutionary rates of symbiotic genes compared to those of other genomic regions not directly implicated in nodulation processes.
| S-EPMC4856229 | biostudies-literature
Project description:ObjectivesWe aim to estimate geographic variability in total numbers of infections and infection fatality ratios (IFR; the number of deaths caused by an infection per 1,000 infected people) when the availability and quality of data on disease burden are limited during an epidemic.MethodsWe develop a noncentral hypergeometric framework that accounts for differential probabilities of positive tests and reflects the fact that symptomatic people are more likely to seek testing. We demonstrate the robustness, accuracy, and precision of this framework, and apply it to the United States (U.S.) COVID-19 pandemic to estimate county-level SARS-CoV-2 IFRs.ResultsThe estimators for the numbers of infections and IFRs showed high accuracy and precision; for instance, when applied to simulated validation data sets, across counties, Pearson correlation coefficients between estimator means and true values were 0.996 and 0.928, respectively, and they showed strong robustness to model misspecification. Applying the county-level estimators to the real, unsimulated COVID-19 data spanning April 1, 2020 to September 30, 2020 from across the U.S., we found that IFRs varied from 0 to 44.69, with a standard deviation of 3.55 and a median of 2.14.ConclusionsThe proposed estimation framework can be used to identify geographic variation in IFRs across settings.
| S-EPMC10990719 | biostudies-literature